Everything you think about Game Mastering is probably wrong: Part II - How to encourage metagaming
In this second installment I will take to task what I consider the biggest misunderstanding of the medium of roleplaying games, the aversion to metagaming.
I am not going to do a big recap of what is metagaming, I assume you know. If your characters run into a person with bite marks in their neck who is faint and has nightmares of someone entering their chamber at nights and the players go "oh, its a vampire, get the garlic and wooden stakes" or if they are crossing a bridge and they are told a "troll leaps forward!" and the wizard switches to fire spells despite never having fought a troll before. Hell, lets go to a higher level of metagaming and go to people sharing information around the table and all collaborating information, even if the characters involved wouldn't know ("oh, I found a medallion in that guys room last week, I bet he has one on his body that would open the chest. I'd check but I am not there"). On the low end some people considering knowledge of intangible aspect of a character (hitpoints, or in some cases exact skill percentages).
You are playing a roleplaying GAME.
The reason GMs keep running into problems with metagaming, is because people like metagaming. It is fun, or it wouldn't keep happening. In no other game would you see an aspect people really like interacting with, and then try to stifle it. You would never go to someone playing scrabble and tell people not to use obscure words or tell them to avoid going for word score multipliers under some form of logic that people should try to use a word with the most base points instead of one with less base points but a higher overall score due to word score multiplier. That would be bizarre.
Metagaming is in the end, players being better at a game than the characters they are playing. So lean into that. One of the recurring themes in how I run games (that I keep thinking I should make explicit in NGR), is that the GM is a referee and the other players are Storytellers. Let the players tell the type of story they want. If you put in the very obvious vampire victim, let them choose if they want to be bumbling idiots who can't put two and two together or if they want to be elite vampire hunters.
The awesomeness mechanic in NGR is there specifically to make both play styles viable, not just on a party level, but on a player by player level. If the person playing Agent Mully WANTS to be the skeptic who believes the local MD's explanation of viral anemia (and Dr. Acula has a very prestigious medical degree, why wouldn't you believe him?) and another is playing Former-Agent Sculder, a seasoned paranormal investigator who knows its a vampire, let both play out. Hell, you can even have the third player just full on metagame and go "Ok, his name is Dr. Acula, I am pretty sure he's the vampire so my character is just going to investigate him, as the random first person in town to look into".
As a GM you might think this is a horrid mess to run and people aren't in the spirit of the game, to which I say "git gud".
Seriously, if you don't want that to play out in exactly that kind of manner then you need to own up to a few things. Why did you make it an obvious vampire symptoms? Why on earth did you make the guy named "Dr. Acula".
There are only three reasons.
1.) You are lazy and using stock templates and cliches right out of central casting but want players to do the work not noticing it and pretending you put in more work than you did. This is the social equivalent of buying a pie from the store, heating it up in your oven and then pretending you made the pie yourself.
2.) Its an intentional red herring because you want to fuck with players with metagame, in which case why do that work of making it a metagame trap and then get mad when players use the adventure in the manner you intended it. This goes perfectly if people metagame and you can all be happy about it and have a good laugh.
3.) You put them in intentionally because you want to use cliches to do the heavy lifting, you WANT people to get the obvious investigation pieces out of the way quickly so they can dive into the deeper mysteries. If you have a dragon in an old dwarf hold, you put it there so people know "yes, its is a greedy miser who sleeps on a giant horde of gold and probably breathes fire" with minimal exposition needed and then they can begin the point of the scenario which is "how do I get the gold and live". In the vampire example you have set it up so obvious who is the vampire because that portion of the investigation isn't the fun part in your scenario, its just the quick set up to the real challenge of "now how do you deal with it without going to prison for murdering a doctor".
Next we discuss to the table talk version of metagaming. One character is stuck in a room with a puzzle they need to solve to exit. Everyone else at the table is making suggestions of what to do or pointing out things they know the character in question may or may not know. Or maybe its not a puzzle, its one person having a lone conversation with one NPC. Is this a problem? No. Same reason.
You are playing a game, a multiplayer social game. I'll go off on a tangent to explain why. One very valid criticism from NGR is that stealth activities are better with multiple characters. This is because of how luck points work, specifically that the group can pool them to keep hidden. Multiple level 0 NPCs without luck points makes the situation worse but multiple heroes (or villains for the murderhobo crowd) makes it easier to infiltrate the enemy castle. This is a bit counter-intuitive in the real world where a stealth mission is easier for a lone operative than a band of 5.
But that is shitty for a social game, especially one where one class is built around stealth. You get into an age old RPG design problem. You either have everyone but one person sitting around watching one person playing their individual stealth/hacking/other minigame for an extended period of time OR you shrink down one person's time to shine down to a couple of quick rolls to get the table back to playing and then the stealth character (or hacker or whatever) goes back to sucking at the non-stealth activities. To this end a lot of games have started to make stealth characters better fighters (high damage glass cannons, DPS is the term I think) and making warriors more of a tank role. I disagree with this too as stealth is a fun strategic thing. Hence NGR's entire stealth system. Why is this relevant to metagaming? Because the one person stuck in a puzzle or one person having a private conversation is boring as shit for the rest of the table. Those things are important to have in adventures, but if they last more than two minutes you shouldn't turn the game into a spectator sport where everyone at the table watches one person play. Let people share information and tactics, let them use their magical telepathic communication channel that all PC's have and which they already use to establish who should be allowed to follow them into a dark hole in search of treasure where they could easily be killed in their sleep for their wealth.
This all leads back to the basic premise. Metagaming is something people enjoy doing in gaming as part of the medium. While there is a type of fun in figuring things out, there is also the fun of putting your vast knowledge of nerd minutia to use and already knowing the right answer. Don't elevate the type of people who enjoy solving new puzzles in an escape room above the type of people who enjoy trivia night at the bar. Throw a mix of both types of problems. The "Age of Myth" campaign I ran specifically leaned into this. You could play and not understand any of the references, but I wanted people who had vast swathes of sci-fi trivia in their brain to be able to make the mental connections to what was going on and what the consequences of various discoveries were.
Knowing that metagaming is part of the game and something people enjoy, make sure you design around it rather than trying to squash it.